Mechanism of inhibition of deoxyribonucleic acid synthesis by 1-. beta

S. H. Gray , C. F. Ainsworth , C. L. Bell , S. S. Danyluk , M. Maccoss ... Effect of ara-C incorporation on deoxyribonucleic acid synthesis in cells ...
0 downloads 0 Views 653KB Size
Biochemistry 1980, 19, 2 15-2 19 Schulman, H., & Greengard, P. (1978) Proc. Natl. Acad. Sci. U.S.A. 75, 5432. Seamon, K. B. (1977) Ph.D. Thesis, Carnegie-Mellon University. Seamon, K. B. (1979) Biochem. Biophys. Res. Commun. 86, 1256. Seamon, K. B., Hartshome, D. J., & Bothner-By, A. A. (1977) Biochemistry 16, 4039. Sin, I. L., Fernandes, R., & Mercola, D. (1978) Biochem. Biophys. Res. Commun. 82, 1132. Snyder, G. H., Rowan, R., Karplus, S., & Sykes, B. D. (1975) Biochemistry 14, 3765. Stevens, F. C., Walsh, M., Ho, H . C., Teo, T. S . , & Wang, J. H. (1976) J . Biol. Chem. 251, 4495. Teo, T. S., Wang, J. H., & Wang, J. C. (1973) J . Biol. Chem. 248, 588. Teshima, Y., & Kakiuchi, S . (1974) Biochem. Biophys. Res. Commun. 56,489. Vanaman, T. C., Sharief, F., & Watterson, D. M. (1977) in Calcium Binding Proteins and Calcium Function (Was-

215

serman, R. H., et al., Eds.) p 107, Elsevier, New York. Van Eerd, J. P., & Kawasaki, Y. (1972) Biochem. Biophys. Res. Commun. 47, 859. Waisman, D., Stevens, F. C., & Wang, J. H. (1975) Biochem. Biophys. Res. Commun. 65, 975. Waisman, D. M., Singh, T. J., & Wang, J. (1978) J . Biol. Chem. 253, 3387. Walsh, M., & Stevens, F. C. (1977) Biochemistry, 16, 2742. Walsh, M., & Stevens, F. C. (1978) Biochemistry 17, 3924. Walsh, M., Stevens, F. C., Oikawa, K., & Kay, C. M. (1978) Biochemistry 17, 3924. Watterson, D. M., Harrelson, W. G., Jr., Keller, P. M., Sharief, F., & Vanaman, T. C. (1976) J . Biol. Chem. 251, 4501. Weeds, A., & McLachlan, A. (1974) Nature (London) 252, 646. Wolff, D. J., Poirier, P. G., Brostrom, C. O., & Brostrom, M. A. (1977) J . Biol. Chem. 252, 4108. Yagi, K., Yazawa, M., Kakiuchi, S., Ohshima, M., & Uenishi, K. (1978) J . Biol. Chem. 253, 1338.

Mechanism of Inhibition of Deoxyribonucleic Acid Synthesis by 1-P-D-Arabinofuranosyladenosine Triphosphate and Its Potentiation by 6-Mercaptopurine Ribonucleoside 5’-Monophosphatet Marietta Y. W. Tsang Lee,* John J. Byrnes,s Kathleen M. Downey,ll and Antero G. So*

ABSTRACT:

The mechanism of inhibition of DNA synthesis by 1$3-D-arabinofuranosyl-ATP (“ATP) and the potentiation of this inhibition by 6-mercaptopurine ribonucleoside 5’-monophosphate (6-MPR-P) have been investigated with mammalian DNA polymerase 6 by using poly(dA-dT) as the template. The inhibition of DNA synthesis by ara-ATP correlates with incorporation of ara-AMP into poly(dA-dT). Nearest-neighbor analysis indicates that ara-AMP does not act as an absolute chain terminator but rather that chains with 3’-terminal arabinosyl nucleotides are extended slowly. The

inhibition of DNA synthesis by ara-ATP is markedly enhanced by the addition of the nucleotide derivative of 6-mercaptopurine, 6-mercaptopurine ribonucleoside 5’-monophosphate. The increased inhibition of DNA synthesis in the presence of 6-MPR-P is due to increased incorporation of ara-AMP. The mechanism by which 6-MPR-P increases the incorporation of ara-AMP is by selective inhibition of the 3’ to 5‘ exonuclease activity of DNA polymerase, thereby preventing the removal of newly incorporated ara-AMP at 3‘ termini of DNA chains.

x e purine nucleoside analogue 1-P-D-arabinosyladenine (ma-A) and the pyrimidine nucleoside analogue 1-&Darabinosylcytosine (ara-C) inhibit the growth of a variety of mammalian cells and viruses. The active metabolites of these nucleoside analogues are their corresponding 5’-triphosphates,

ara-ATP and ara-CTP, which are potent inhibitors of DNA synthesis in vitro. Kinetic studies have shown that ara-ATP and ara-CTP are competitive inhibitors of DNA synthesis when the corresponding deoxynucleoside triphosphate, dATP or dCTP, is the variable substrate (Furth & Cohen, 1968; Muller et al., 1975). However, the exact mechanism by which ara-CTP inhibits DNA synthesis remains uncertain. Studies with partially purified DNA polymerases have shown that nearly all of the incorporated ara-CMP is present at the 3’hydroxyl termini of DNA chains (Momparler, 1969, 1972; Waqar et al., 1971). These findings led to the proposal that ara-C inhibits DNA synthesis by acting as a chain terminator, either preventing or markedly slowing the addition of deoxynucleotides (Momparler, 1969, 1972; Waqar et al., 1971; Magnusson et al., 1974; Burgoyne, 1974). However, studies with intact cells and cell lysates have indicated that ara-CMP and ara-AMP are incorporated primarily in internucleotide linkage of DNA. It was also found that inhibition of DNA synthesis by low concentrations of ara-C is readily reversible in vivo; i.e., removal of the inhibitor leads to the rapid re-

t From the Howard Hughes Medical Institute Laboratory, Departments of Medicine, Biochemistry, and Oncology, and the Center for Blood Diseases, University of Miami School of Medicine, and the Department of Medicine, Veterans Administration Hospital, Miami, Florida 33101. Received August 18, 1979. This research was supported by grants from the National Institutes of Health (NIH AM 09001 and AM 26206) and the National Science Foundation (PCM 77-17651 1) and in part by a grant to the Comprehensive Cancer Center to the University of Miami from the National Cancer Institute (NIH-CA14395) and by research funds given in memory of Mary Beth Weiss and Elizabeth Wenig. *Correspondence should be addressed to this author at the Department of Medicine, University of Miami School of Medicine, Miami, FL. He is an Investigator of the Howard Hughes Medical Institute. 8 Research Associate, Howard Hughes Medical Institute. Clinical Investigator, Veterans Administration. 11 Research Career Development Awardee (NIH-K04-HL-0003 1).

0006-2960/80/0419-0215$01 .OO/O

0 1980 American Chemical Society

216

BIOCHEMISTRY

sumption of DNA replication a t a normal rate. These results, together with the observations that ara-ATP and ara-CTP are competitive inhibitors of DNA synthesis, led to the suggestion that the inhibition of DNA synthesis by arabinosyl nucleotides is not due to their incorporation into DNA but rather to inhibition of DNA polymerase activity (Furth & Cohen, 1968; Graham & Whitmore, 1970; Hunter & Francke, 1975; Muller et al., 1975; Manteuil et al.. 1974; Wist et al., 1976; Dicioccio & Srivastava, 1977). The effectiveness of ara-C as a chemotherapeutic agent in the treatment of acute myelogenous leukemia is significantly enhanced when used in combination with the purine analogues 6-thioguanine (6-TG) or 6-mercaptopurine (6-MP) (Gee et al., 1969; Carey, 1970; Buchanan et al., 1973). The mechanism of this synergism is unknown. We have recently shown that the active metabolite of 6-MP, 6-mercaptopurine ribonucleoside 5’-monophosphate (6-MPR-P), selectively inhibits the 3’ to 5’ exonuclease activity associated with mammalian DNA polymerase 6 while the DNA polymerase activity is not inhibited (Byrnes et al., 1976, 1977). Since the proposed function of the 3’ to 5’ exonuclease is to excise mismatched nucleotides or nucleotide analogues incorporated at the primer terminus during DNA synthesis (Brutlag & Kornberg, 1972; Bessman et al., 1974), the selective inhibition of the proofreading exonuclease activity by 6-MPR-P might be expected to result in an increased net incorporation of nucleotide analogues. In this study we have investigated the effects of metabolites of ara-A and 6-MP on DNA synthesis using purified DNA polymerase 6 and poly(dA-dT). We shall present data showing that (1) the inhibition of DNA synthesis by ura-ATP is a consequence of ara-AMP incorporation into the 3’hydroxyl termini of DNA chains, ( 2 ) the inhibitory effect of ara-ATP is potentiated by 6-MPR-P, (3) the potentiation of ara-ATP inhibition of DNA synthesis by 6-MPR-P is due to an increased net incorporation of ara-AMP into DNA, and (4) the mechanism by which 6-MPR-P increases incorporation of ara-AMP is by selective inhibition of the 3’ to 5’ exonuclease activity of DNA polymerase 6, thereby preventing hydrolysis of incorporated nra-AMP. Materials and Methods 3H-Labeled deoxyribonucleoside triphosphates were obtained from New England Nuclear. [3H]-ara-ATP was custom synthesized by ICN, and its purity was examined by thin-layer chromatography on poly(ethy1enimine)-cellulose (PEIkellulose) as described by Cashel et al. (1969). Less than 1.5% contamination by nucleoside diphosphates or monophosphates was detected. To rule out the possibility that [3H]-ara-ATPwas contaminated by [3H]dATP, [3H]-araATP was enzymatically converted to [3H]-ara-AMP by the template-dependent conversion of deoxyribonucleoside triphosphate to monophosphate reaction, catalyzed by purified D N A polymerase 6 with poly(dA-dT) as the template. The products of the reaction were examined by PEI-cellulose thin-layer chromatography as described by Waqar et al. (1971), and no contamination by 13H]dAMP was detected ( C 1%). Unlabeled deoxyribonucleoside triphosphates were obtained either from P-L Biochemicals or Calbiochem. The barium salt of 6-mercaptopurine ribonucleoside 5’-monophosphate was obtained from P-L Biochemicals and converted to the sodium salt before use. Poly(dA-dT) was obtained from Biopolymers, Inc., and was dialyzed against two changes of 0.06 M KCl and 0.01 M Tris-HCI buffer, pH 7.4 (DNA buffer), before use. PEI-cellulose thin-layer chromatography plates were purchased from Brinkman Instruments, Inc., and were prerun with distilled water and stored at 0 “ C before use.

LEE, B Y R h E S , D O W N E Y , A N D SO

Salmon sperm DNA, micrococcal nuclease, and bovine spleen phosphodiesterase I1 were obtained from Worthington Biochemical Corp. Preparation of Enzyme. DNA polymerase 6 (step VI) was prepared from rabbit erythroid hyperplastic bone marrow as previously described (Byrnes et al., 1976). One unit of DNA polymerase catalyzes the incorporation of 1 nmol of d T M P in 1 h at 37 “C. Preparation of 3’- Terminus [3H]-ara-AMP-Labeled Poly(dA-dT). Poly(dA-dT) labeled a t the 3’ terminus with [3H]-ara-AMPwas prepared by incubating poly(dA-dT) with DNA polymerase 6 and [3H]-nra-ATP. The reaction mixture contained in a final volume of 2.5 m L the following: 66 mM Hepes (N-2-hydroxyethylpiprazine-N’-2-ethanesulfonic acid) buffer, pH 6.5; 0.79 A260unit of poly(dA-dT); 66 m M KCI; 0.13 m M MnCI,; 1.2 mM 6-MPR-P; 1.5 p M [3H]-ara-ATP, 6800 cpm/pmol; 3.4 units of DNA polymerase 6. The reaction mixture was incubated at 37 “C for 15 min and stopped by the addition of 12.5 pmol of EDTA (ethylenediaminetetraacetic acid), 1.25 mmol of sodium perchlorate, and sodium dodecyl sulfate to a final concentration of 1% (w/w). The solution was deproteinized twice with chloroform-isoamyl alcohol (24:1), and poly(dA-dT).[3H]-ara-AMPwas separated from unincorporated nucleotides by dialysis against two changes of DNA buffer. The final specific activity of poly(dA-dT).[3H]-ara-AMPwas 6.34 X lo4 cpm/A260unit. DNA Polymerase Assay. Each reaction mixture contained in a final volume of 0.25 mL the following: 80 m M Hepes buffer, pH 7.0; 125 m M KCI; 0.6 m M MnCI,; 25 p M poly(dA-dT); 0.5 unit of DNA polymerase 6. The concentration of the deoxynucleoside triphosphate substrates is given for each experiment. After 15 min a t 37 “C the reaction was terminated by the addition of 2 mL of cold 5% trichloroacetic acid and 20 pL of 20 mM dATP. The precipitate was collected, washed, and counted as previously described (Byrnes et al., 1976). Results Correlation of Inhibition of DNA Synthesis with Incorporation of ara-AMP into DNA. The effects of increasing concentrations of [3H]-ara-ATPboth on [3H]-ara-AMPincorporation and on poly(dA-dT) synthesis, as measured by the incorporation of [32P]dTMP, are shown in Figure 1. Increasing concentrations of [3H]-am-ATP result in increased [3H]-ara-AMPincorporation but decreased [32P]dTMPincorporation into poly(dA-dT). The correlation between the degree of inhibition of D N A synthesis and the amount of ara-AMP incorporated into DNA suggests that the inhibition of DNA synthesis by ara-ATP is the result of its incorporation into poly(dA-dT). Potentiation of ara-ATP Inhibition of DNA Synthesis by 6-MPR-P. The effects of 6-MPR-P on the inhibition of poly(dA-dT) synthesis by ara-AMP is shown in Figure 2. In the absence of 6-MPR-P, poly(dA-dT) synthesis is inhibited 50% at an ara-ATP concentration of 4 X 10” M , whereas in the presence of 8 X M 6-MPR-P, the concentration of ara-ATP required to inhibit 50% is reduced to 1.5 X 10” M. Higher 6-MPR-P concentrations ( 1 . 2 X M) resulted in a further reduction in the concentration of ara-ATP required to inhibit poly(dA-dT) synthesis. Thus, the inhibition of DNA synthesis by ara-ATP is potentiated by 6-MPR-P. No inhibition of D N A synthesis is observed in the presence of 6MPR-P alone (data not shown). Effect of 6-MPR-Pon Incorporation of [3H]-ara-AMP into Poly(dA-do. To determine whether the potentiation of ara-ATP inhibition of DNA synthesis by 6-MPR-P is due to

V O L . 1 9 , N O . 1,

INHIBITION OF DNA SYNTHESIS BY ARA-ATP

Y

I

I

I

I

I

I

2

4

6

8

IO

12

[ARA-ATP]

I

I

I

I

I

I

I

02

04

06

08

IO

12

4M

[6-MPR-P]

Effect of ara-ATP concentration on [3H]-ara-AMPand [32P]dTMPincorporation. Assays were carried out as described under Materials and Methods exce t that the KC1 concentration was 45 mM. The concentration of [ ( U - ~ ~ P I ~was T T4.6 P pM, 1707 cpm/pmol, dATP was present at 0.4 pM, and the concentration of ['HI-ara-ATP (1917 cpm/pmol) was varied from 0 to 11.2 p M , The samples were counted in toluene-Omnifluor scintillant in a Packard Model 33AT liquid scintillation counter. The I2P spillover was 0.7% and the 'H spillover was less than 0.04%. FIGURE 1:

--

2

4 6 [ IRb,ATP] FM

8

1

FIGURE 2: Effect of 6-MPR-P on the inhibition of DNA synthesis by ara-ATP. Assays were carried out as described under Materials and Methods. The concentration of ['HIdATP was 0.45 pM, 7787 cpm/pmol, and the concentration of dTTP was 4 pM. ara-ATP was M 6-MPR-P (a); varied as indicated. No 6-MPR-P (0);8 X 1.2 X M 6-MPR-P (w).

an increased incorporation of ara-AMP into poly(dA-dT), we have investigated the effects of 6-MPR-P on the incorporation of ['HI-ara-AMP into poly(dA-dT) (Figure 3). Increasing concentrations of 6-MPR-P increase the incorporation of [3H]-ara-AMP into poly(dA-dT); the incorporation of [3H]-ara-AMPis increased 1 1-fold at 1 X lo4 M and 14-fold at 4 X M 6-MPR-P. These results are also consistent with the proposal that the inhibition of poly(dA-dT) synthesis by ara-ATP is due to its incorporation into DNA. Increased Incorporation of ara-AMP is Due to Inhibition of 3'to 5'Exonuclease Activity of DNA Polymerase 6. We have previously shown that the 3' to 5' exonuclease activity associated with both mammalian and bacteria DNA polymerases can be selectively inhibited by nucleoside 5'-monophosphates, whereas the DNA polymerase activity is not affected (Byrnes et al., 1977; Que et al., 1978). Thus, it is likely that the increase in the incorporation of ['HI-nra-AMP into poly(dA-dT) in the presence of 6-MPR-P is due to the inhibition of hydrolysis of newly incorporated [ 3H]-ara-AMP. That this is the case is seen in Figure 4, which shows the effects of increasing concentrations of 6-MPR-P both on the incor-

mM

I

I

04

06 [&MPR-P]

0

217

FIGURE 3: Effect of 6-MPR-P on the incorporation of ['HI-ara-ATP into DNA. Assay conditions were as described under Materials and Methods. The concentration of ['HI-ara-ATP was 1.7 pM, 4230 cpm/pmol; 6-MPR-P was present as indicated.

02

01

1980

08

10

12

mM

FIGURE 4: Effect of 6-MPR-P on the incorporation and hydrolysis of ['HI-ara-AMP. The reaction mixtures were as described under Materials and Methods except that the KCI concentration was 85 mM and the poly(dA-dT) concentration was 6 FM. The [3H]-araATP concentration was 1.7 pM, 4230 cpm/pmol. The reaction was incubated at 37 "C for 15 min and stopped by rapid chilling and the addition of 20 pL of 0.1 M EDTA. An aliquot of the reaction mixtwe (0.1 mL) was pipetted into a tube containing 2 mL of 5% trichloroacetic acid. The precipitate was collected, washed, and counted as previously described (Byrnes et al., 1976). A second aliquot of the reaction mixture (0.12 mL) was spotted on a thin-layer plate marked into 1.2-cm lanes. The plate was predeveloped with absolute methanol, and, after drying, the plate was developed with 1 N acetic acid to 3 cm above the origin, followed by 0.3 M LiCl (Randerath & Randerath, 1965). Each lane was subsequently cut into 18 I-cm sections and counted. ara-ATP and ara-AMP standards were run in parallel as markers and identified by ultraviolet absorption at 254 nm. Minus enzyme, minus template, and minus incubation controls were run with the experimental assays.

poration of ['HI-ara-AMP into polynucleotide and on the template-dependent conversion of ['HI-ara-ATP to [3H]ara-AMP. The template-dependent generation of ['HI-araAMP represents the successive incorporation and hydrolysis of [3H]-ara-AMP at the 3'-hydroxyl termini of the template/primer by the DNA polymerase and 3' to 5' exonuclease activities of DNA polymerase 8. Increasing concentrations of 6-MPR-P result in decreased hydrolysis of newly incorporated ['HI-ara-AMP and, thus, in an increased net incorporation of [3H]-ara-AMP into poly(dA-dT). It is interesting to note that, when the exonuclease is active, most of the ara-AMP that is incorporated is hydrolyzed and released as free ara-AMP, suggesting that most of the araAMP that has been incorporated remains at 3' termini and

218

L E E , B Y R N E S , DOWNEY, A N D S O

BIOCHEMISTRY

Table I: Transfer of 32Pfrom [e3’P]dTTP toara-AMP or dAMP

substrates dATP, [ 32P]dTTP dATP, [ 32P]dTTP ara-ATP, [”PI dTTP ara-ATP, [ 32P]dTTP ara-ATP, [ 32P]dTTP, dATP ura-ATP, [”PI dTTP, dATP 2

1

6

8

10

INCUBATION TIME (min)

FIGURE 5 :

Effect of 6-MPR-P on the hydrolysis of terminally labeled poly(dA-dT)-[-’H]-ara-AMP. Each reaction mixture contained in a final volume of 0.15 mL the following: 133 mM Tris-HC1, pH 7.8; 6.34 X lo4 ~ p m / A , , ~ 0.1 A260 unit of p~ly(dA-dT)-[~H]-ara-AMP, unit; 0.33 mM MnC1,; 0.5 unit of DNA polymerase d; 2 mM 6-MPR-P as indicated. The reaction was incubated at 37 OC for the indicated time and stopped by the addition of 2 mL of cold 5% C l & O H . After the mixture was allowed to stand for 10 min at 0 OC, the mixture was filtered on Millipore HAWP filter disks and washed with 30 mL of distilled water. The disks were dried and counted as previously described (Byrnes et al., 1976). Similar results were obtained when assayed under D N A polymerase assay conditions in the absence of deoxyribonucleoside triphosphates (unpublished observation).

is not extended. This is consistent with the hypothesis that ara-ATP inhibition of DNA synthesis is a consequence of its incorporation into DNA and that incorporation of ara-AMP slows the rate of DNA chain extension. Also consistent with this hypothesis is the observation that inhibition of the 3’ to 5’ exonuclease activity results in a 10-fold inhibition of araAMP turnover but only a 2-fold stimulation of its incorporation into poly(dA-dT). If the incorporation of ara-AMP at 3’ termini slows the rate of further DNA chain growth, then the rate of ara-AMP incorporation would be limited by the availability of template/primer (which is limiting in the present case) whereas the generation of free ara-AMP would not be dependent on the availability of template/primer since the hydrolysis of 3’-terminal ara-AMP regenerates effective 3’ termini. To further demonstrate that the inhibition of 3’ to 5’ exonuclease activity by 6-MPR-P is the mechanism responsible for increased incorporation of ara-AMP into DNA, the effect of 6-MPR-P on the hydrolysis of 3’-terminally labeled poly(dA-dT).[3H]-ara-AMP by the 3’ to 5’ exonuclease activity of DNA polymerase 6 was investigated (Figure 5 ) . In the absence of 6-MPR-P, the rate of release of [3H]-ara-AMP is very rapid; approximately 90% is released after 2 min of incubation and complete removal is attained after 5 min. However, in the presence of 6-MPR-P very little of the p~ly(dA-dT).[~H]-ara-AMP is hydrolyzed; after 10 min of incubation only 20% of the [3H]-ara-AMPhad been released. Extension of Chains with 3’-Terminal ara-AMP. To determine whether DNA polymerase 6 can extend DNA chains with 3’-terminal ara-AMP, we have analyzed the enzymatic digest of poly(dA-dT) synthesized in the presence of ara-ATP and [ ( u - ~ ~ P I ~ T(Table T P I). The transfer of 32Pto ara-AMP would indicate that ara-AMP is incorporated into internucleotide linkage and that chains with 3’-terminal arabinosyl nucleotides can serve as primers for chain extension. In the absence of 6-MPR-P, very little 32Poriginally present in dTMP is transferred to urn-AMP. However, in the presence of 6MPR-P, which inhibits the 3’ to 5’ exonuclease activity of DNA polymerase 6, the transfer is stimulated fourfold. The

[”PI6ara-AMP [32P]dAMP MPR-P (pmol) @mol) -

0 0

-

-

1.1 4.0 6.5

38 152 1.6 6.0 28

+

15.8

85

+ +

a Assay conditions were as described under Materials and Methods except that the concentration of KCI was 105 mM and 3.08 X A , , , unit of poly(dA-dT) was used. The concentration of [Q-~’P]~TTPwas 0.6 p M (900 cpm/pmol). The concentrations of ara-ATP and dATP were 4 pM. The concentration of 6-MPR-P, when present, was 1.2 X l o v 3M. The mixtures were incubated at 37 “C for 60 min and terminated by the addition of 0.5 mL of 20% trichloroacetic acid, 50 pL of Salmon sperm DNA, 2 mg/mL, and 0.1 mL of 0.3 M sodium pyrophosphate. Samples were hydrolyzed by the procedure of Wu & Kaiser (1968). The DNA product was washed 3 times by repeated suspension and precipitation, extracted 3 times with ether, and dried under vacuum. The dried DNA was redissolved in 0.3 mL of 0.1 M Tris-acetate buffer, pH 8.4, and digested by the addition of 10 pL of 0.05 M CaCl, and 20 pL of micrococcal DNase (500 pg/mL), followed by incubation for 1 h at 37 “C. The incubation mixture was adjusted to pH 6.4 by the addition of 2 pL of 0.2 M acetic acid and 10 pL of 0.1 M Tris-acetate, pH 6.4. The reaction mixture was also made 1.5 mM in potassium phosphate, pH 6.4, to inhibit phosphatase activity. The mixture was lyophilized to dryness, 50 pL of H,O and 20 p L of spleen phosphodiesterase, 11 units/mL, were added, and the mixture was incubated for 90 min at 37 “C. The mixture was then heated at 100 “ C for 3 min. Samples and standards were spotted on PEI-cellulose thin-layer plates and developed twice in isobutyric acid-1 M NH,OH-0.1 M EDTA (100:60:1 by volume) (Waqar et al., 1971). TheRf of dAMP was 0.76 and that ofaru-AMP was 0.59.

transfer of 32Pto ara-AMP is further stimulated by the addition of dATP. In the presence of both dATP and ara-ATP, the transfer of 32Pto ara-AMP is only 20% of that transferred to dAMP, suggesting that, although DNA primers with 3’terminal arabinosyl nucleotides can serve as sites for chain elongation, they are inefficient primer termini compared to chains with 3’-terminal deoxyribosyl nucleotides. The amount of 32Ptransferred to dAMP when no dATP was present in the reaction mixture represents transfer of 32Pto dAMP present at the primer termini of the poly(dA-dT) primer/template. Discussion Although it has been over a decade since it was first shown that ara-C and ara-A are potent inhibitors of DNA synthesis, there is still a lack of unanimity as to the mechanism of inhibition of DNA synthesis by arabinosyl nucleotides. There are two main hypotheses. Momparler (1968) has suggested that inhibition of DNA synthesis is a consequence of the incorporation of arabinosyl nucleotides into 3’ termini of DNA chains, thereby preventing further chain extension. This hypothesis is based upon in vitro results obtained with a partially purified DNA polymerase. A modified version of this hypothesis which is gaining wider acceptance suggests that arabinosyl nucleotides do not act as absolute chain terminators, but rather the incorporation of arabinosyl nucleotides at 3’ termini markedly slows the addition of subsequent deoxynucleotides and this is the basis of inhibition of DNA synthesis (Magnusson et al., 1974; Burgoyne, 1974; Cozzarelli, 1977). The other hypothesis is that the direct inhibition of DNA polymerase activity and not the incorporation of arabinosyl

I N H I B I T I O N O F D N A S Y N T H E S I S BY A R A - A T P

nucleotides into DNA is the cause of inhibition of DNA synthesis. This hypothesis is based upon results of studies with intact cells in which it was found that (1) there is no correlation between the amount of arabinosyl nucleotide incorporated and the degree of lethality induced by this analogue, (2) most of the ara-CMP incorporated into DNA is found in internucleotide linkage and only a small percentage is located at 3’ termini, and (3) inhibition of DNA synthesis is readily reversible-removal of the inhibitor leads to rapid resumption of DNA replication at normal rate. It was assumed that incorporation of arabinosyl nucleotides into DNA would result in irreversible inhibition and that no mechanism exists in the cell to excise nucleotide analogues blocking DNA chain extension (Graham & Whitmore, 1970; Wist et al., 1976). Previous interpretation of the above data has not taken into consideration the possibility that mammalian DNA polymerases might be similar to microbial enzymes in being associated with a proofreading 3’ to 5’ exonuclease activity whose function is to remove mismatched nucleotides or nucleotide analogues that are incorporated during DNA synthesis (Brutlag & Kornberg, 1972; Bessman et al., 1974; Byrnes et al., 1976). The observations with intact cells that the amount of ara-CMP incorporated into DNA is very small and that it is present primarily in internucleotide linkage may be due to preferential hydrolysis of 3‘-terminal arabinosyl nucleotides by the 3’ to 5’ exonuclease activity of mammalian DNA polymerase during the isolation of the cells. Similarly, the reversibility of the inhibition of DNA synthesis by arabinosyl nucleotides may also be explained by the efficient hydrolysis of incorporated arabinosyl nucleotides at the 3’ termini of growing DNA chains. The present study indicates that inhibition of DNA synthesis by ara-ATP is consequent to incorporation of ara-AMP into DNA and not to inhibition of DNA polymerase activity. This is based on the finding that there is a correlation between the degree of inhibition of DNA synthesis and the amount of ara-AMP incorporated into DNA. Increasing concentrations of ara-ATP result in increased incorporation of ara-AMP but decreased incorporation of dTMP. We have further demonstrated that ma-AMP is not an absolute chain terminator and that ara-AMP is incorporated into internucleotide linkage in DNA chains, since 32Pcan be transferred from [32P]dTMPto ara-AMP with poly(dA-dT) as primer/template. However, the small amount of 32P transferred to ara-AMP as compared to dAMP is consistent with the hypothesis that chains terminated with arabinosyl nucleotides are extended very slowly (Magnusson et al., 1974; Burgoyne, 1974; Cozzarelli, 1977). We have observed that 6-MPR-P potentiates the inhibition of DNA synthesis by ara-ATP. We have demonstrated that this potentiation is due to increased net incorporation of ara-AMP into the 3’ termini of poly(dA-dT) and that selective inhibition of the 3’ to 5’ exonuclease activity of DNA polymerase 6 by 6-MPR-P is the mechanism by which 6-MPR-P increases incorporation of ara-AMP into DNA. The demonstration that selective inhibition of the 3’ to 5’ exonuclease activity of DNA polymerase by 6-MPR-P prevents the removal of arabinosyl nucleotides incorporated at the 3’ termini of DNA chains and that this results in increased inhibition of DNA synthesis is also consistent with the hypothesis that inhibition of DNA synthesis by ara-ATP is due to the incorporation of arabinosyl nucleotides into DNA. Studies on the metabolism of purine analogues in animal models suggest that the in vivo concentrations of 6-MPR-P may be sufficient to inhibit the exonuclease activity of DNA

V O L . 19, N O . 1 , 1980

219

polymerase. 6-MP and 6-TG are good substrates for adenylic pyrophosphorylase and inosinic, guanylic pyrophosphorylase, respectively, and are readily converted to the nucleoside 5’monophosphates. Since 6-MPR-P and 6-TGR-P are poorly converted to the diphosphate or triphosphate, the concentration of the nucleoside 5’-monophosphate derivatives of these purine analogues can reach very high levels in vivo, i.e., greater than M (Paterson, 1959; Moore & LePage, 1958). At this concentration 6-MPR-P inhibits the 3’ to 5’ exonuclease activity of mammalian DNA polymerase 6 (Byrnes et al., 1977). The enhanced effectiveness of ara-C in the treatment of acute myelogenous leukemia when used in combination chemotherapy with 6-TG or 6-MP is well established. The present studies provide a mechanism to account for this synergism and a rational basis for this combination of chemotherapeutic agents. References Bessman, M. J., Muzyczka, N., Goodman, M. F,, & Schnaar, R. L. (1974) J. Mol. Biol. 88, 409. Brutlag, D., & Kornberg, A. (1972) J . Biol. Chem. 247, 241. Buchanan, J. G., Mathews, J. R. D., Davidson, J. G., Grigor, R. R., & Casey, T. P. (1973) Cancer (Philadelphia) 32, 789. Burgoyne, L. A. (1974) Biochem. Pharmacol. 23, 1619-1627. Byrnes, J. J., Downey, K. M., Black, V. L., & So, A. G., (1976) Biochemistry IS, 2817. Byrnes, J. J., Downey, K. M., Que, B., Lee, M., Black, V., & So, A. G. (1977) Biochemistry 16, 3740-3746. Carey, R. W. (1970) Proc. Am. Assoc. Cancer Res. 11, 15. Cashel, M., Lazzarini, R. A., & Kalbacher, B. (1969) J . Chromatogr. 40, 103. Cozzarelli, N. R. (1977) Annu. Rev. Biochem. 46, 641-668. Dicioccio, R. A., & Srivastava, B. I. S. (1977) Eur. J . Biochem. 79,411. Furth, J. J., & Cohen, S . S . (1968) Cancer Res. 28, 2061. Gee, T. S . , Yu, K.-P., & Clarkson, B. D. (1969) Cancer (Philadelphia) 23, 10 19-1032. Graham, F. L., & Whitmore, G. F. (1970) Cancer Res. 30, 2636. Hunter, T., & Francke, B. (1975) J . Virol. IS, 759-775. Magnusson, G., Craig, R., Narkhammer, M., Reichard, P., Staub, M., & Warner, H. (1974) Cold Spring Harbor Symp. Quant. Biol. 34, 277-233. Manteuil, S . , Kopecka, H., Caraux, J., Ariel, P., & Girard, M. (1974) J . Mol. Biol. 90, 751-756. Momparler, R. L. (1968) Biochem. Biophys. Res. Commun. 32, 23. Momparler, R. (1972) Mol. Pharmacol. 8, 362-370. Moore, E. C., & LePage, G. A. (1958) Cancer Res. 18,1075. Muller, W., Rhode, H., Beyer, R., Maidhof, A,, Lachmann, M., Taschner, H., & Zahn, R. K. (1975) Cancer Res. 35, 2160-2168. Paterson, A. R. P. (1959) Can. J . Biochem. Physiol. 37, 1011. Que, B. G., Downey, K. M., & So, A. G. (1978) Biochemistry 17, 1603. Randerath, E., & Randerath, K. (1965) Anal. Biochem. 12, 83. Stenstrom, M., Edelstein, M., & Grisham, J. W. (1974) Exp. Cell Res. 89, 439. Waqar, M., Burgoyne, L., & Atkinson, M. (1971) Biochem. J . 121, 803-809. Wist, E., Krokan, H., & Prydz, H. (1976) Biochemistry 15, 3647-3652. Wu, R., & Kaiser, A. D. (1968) J . Mol. Biol. 35, 523-537.